Ben Smith on the Semafor World Economy Summit: AI anxiety, labor market self-censorship, and the disconnect between private and public sentiment

Ben Smith, Semafor

Ben Smith, co-founder and editor-in-chief of Semafor, is speaking from the Semafor World Economy Summit, a five-day gathering of roughly 500 global CEOs and political figures from the US and Europe running three simultaneous stages.

“There's a real difference between what people say in public and what they say in private. In our kind of Chatham House rule gatherings, you're hearing a lot more concern about a bit about the long term direction of US power, The US economy, The US markets, people hedging away, and no one's saying that stuff in public.”

Every conversation winds up coming back to AI

The dominant thread across the summit is anxiety about AI's effect on large enterprises. Smith says the consensus view is that big, brand-name American companies face the greatest exposure in the transition, and the people who work for, advise, or compete with them are thinking about little else.

On labor, the public numbers look fine, but the mood is darker. There's a broad perception among the American public that AI exists to eliminate jobs, and that hostility is hanging over the conference. Smith flags a specific concern about the class of 2026 — graduates who are native to AI tools and the most qualified to use them, but entering a job market that may not want them. The anxiety runs in two directions: what happens to the workforce if AI eliminates roles, and what happens if the people most capable of deploying these tools can't find work.

Washington's response is minimal. Every member of Congress is fielding questions about AI and labor, but the institutional track record on tech regulation — twenty years of social media hearings with little to show for it — doesn't inspire confidence that federal action is coming. Smith does note one unexpected data point from the conference floor: a call center operator focused on financial fraud reporting that AI-driven fraud has actually increased workload rather than eliminated it, keeping headcount intact.

Self-censorship gap

The sharpest observation from Smith is the gap between what people say in public and what they say in private. In Chatham House-rules settings, Smith is hearing significant concern about the long-term direction of US power, the US economy, and US markets — with executives quietly hedging their exposure. None of it is surfacing publicly. The S&P is at a record high despite over a month of conflict with Iran, and the visible mood at the conference tracks the markets. The private conversations are a different story.

Newsroom tooling

Semafor has built a transcript tool for the summit that ingests every interview in real time and surfaces what's new, flagging genuine news for generalist reporters covering stages they weren't on. Smith also mentions building a personal tool he calls "pre-dunk" — a Google Apps Script that predicts which stories will attract critical tweets — though he acknowledges adoption among his team has been limited.

The broader Semafor editorial bet is that readers want human connection and authentic voice from journalists, not AI-generated content. AI runs behind the scenes to amplify journalists; it stays out of the byline.